Developmental corpus of Slovene (without language corrections) Šolar-Clear

Name: Developmental corpus of Slovene (without language corrections) Šolar-Clear
License: https://creativecommons.org/licenses/by-nc-sa/4.0/

Avtorji: Rozman, Tadeja ; et al.prikaži vse
Rozman, Tadeja ; Stritar Kučuk, Mojca ; Kosem, Iztok ; Krek, Simon ; Krapš Vodopivec, Irena ; Arhar Holdt, Špela ; Stabej, Marko ; Laskowski, Cyprian ; Klemenc, Bojan

Opis: Šolar-Clear is an adapted version of the Šolar 1.0 corpus, cf. http://hdl.handle.net/11356/1036. The Šolar(-Clear) corpus consists of texts written by students in Slovene primary and secondary schools. School essays form the majority of the corpus (64.2%) while other material includes texts created during lessons, such as text recapitulations or descriptions, examples of formal applications etc. Unlike the original Šolar corpus, Šolar-Clear only includes student texts while language corrections and other types of feedback from the teachers are not included. The corpus can thus be used for processing tasks where the inclusion of corrections hinders or complicates the procedures (e.g. for comparative data extraction, training of language models etc).